Policy Platform

Control of Neglected Tropical Diseases in Burundi:Partnerships, Achievements, Challenges, and LessonsLearned after Four Years of Programme ImplementationOnesime Ndayishimiye1, Giuseppina Ortu2*, Ricardo J. Soares Magalhaes3, Archie Clements3,

Johan Willems4, Jane Whitton2, Warren Lancaster5, Adrian Hopkins6, Alan Fenwick2

1 Programme National Integre de lutte contre les Maladies Tropicales Negligees et la Cecite (PNIMTNC) - Ministere de la Sante Publique et de la lutte contre le SIDA,

Bujumbura, Burundi, 2 Schistosomiasis Control Initiative - Imperial College, Department of Infectious Disease Epidemiology, London, United Kingdom, 3 University of

Queensland, Infectious Disease Epidemiology Unit, School of Population Health, Herston, Queensland, Australia, 4 CBM Regional Office for Central Africa, Parklands,

Nairobi, Kenya, 5 END Fund/Geneva Global, London, United Kingdom, 6 Mectizan Donation Programme, Decatur, Georgia, United States of America

Introduction

Neglected tropical diseases (NTDs) are

a group of 17 conditions that together

affect over 1,000,000,000 people in the

developing world [1]. Nongovernmental

organisations (NGOs) and public-private

partnerships have joined efforts to place

the NTDs in the spotlight, especially the

seven that can be treated with an annual

dose of safe and effective drugs: trachoma,

three soil-transmitted helminthiases (STHs,

ascariasis, hookworm infections, and trich-

uriasis), lymphatic filariasis (LF), onchocer-

ciasis, and schistosomiasis (SCH) [2]. On

the wave of this new interest in NTDs,

Burundi was selected in 2007 by the

Legatum Foundation for financial support

for a national NTD programme.

Before 2007, Burundi hosted two min-

isterial programmes related to treatment of

parasitic diseases: the National Pro-

gramme for the Control of Transmissible

Diseases and Deficiencies (Lutte contre le

Maladies Transmissibles et Carentielles

[LMTC]), responsible for managing ma-

laria, worm infections, and nutrition, and

an onchocerciasis control programme

(Programme National de Lutte contre

l’Onchocercose [PNLO]). The PNLO,

supported by the African Programme for

Onchocerciasis Control (APOC) [3] and

CBM (formerly known as Christian Blind

Mission) [4], involved the annual admin-

istration of ivermectin (IVM) in ten

endemic districts (namely Cibitoke, Bu-

banza, Bururi, Rutana, Mpanda, Ru-

monge, Matana, Gihofi, Mabayi, and the

Kayogoro commune in the Makamba

district) via the well-established Commu-

nity Directed Treatment with Ivermectin

(CDTI) strategy [5]. For other NTDs, such

as trachoma and LF, no specific pro-

grammes were in place in Burundi at that

time.

Due to limited funding, the LMTC priori-

tised malaria control activities. Nevertheless,

distribution of praziquantel (PZQ) in

health centres and mebendazole (MBZ)

in schools was already happening between

1973 and 1992 in areas considered highly

endemic [6,7]. Historical epidemiological

data on STHs [8] and a survey performed

in 2005 [9] suggested that both Schistosoma

mansoni and STH infections were wide-

spread, with SCH prevalence varying

across provinces (range 0–61%, mean

22%) and the overall prevalence of STHs

reaching 59.7%.

With the beginning of the civil war in

1993, national prevention and control

activities slowed down considerably. De-

spite the continued political instability, the

Ministry of Health (MoH), with UNICEF

and the WHO, introduced an annual

Mother and Child Health Week (MCHW)

in 2003, during which vaccinations, mi-

cronutrients, bed nets, and deworming

drugs were distributed.

Prior to 2007, trachoma and LF prev-

alence had never been thoroughly inves-

tigated in Burundi. A few cases of

trachoma were suspected among refugees,

and there was evidence of cases of

elephantiasis, one of the clinical manifes-

tations of LF, although they were likely

due to podoconiosis [10].

Sparse historical epidemiologic data and

the presence of a few early-stage health

initiatives (LMTC, PNLO, and MCHW)

created an excellent opportunity to roll out an

NTD control programme in Burundi. With

the specific objectives of defining the popu-

lation at risk for each infection and elaborat-

ing and implementing a drug treatment

strategy targeting the population at risk, the

final aspiration was to create a solid frame-

work for a sustainable NTD control pro-

gramme integrated with the health initiatives

already in place that would continue after the

end of the donor-funded programme.

The present paper provides a critical

overview after four years of a programme

that started in Burundi in June 2007.

Strategies, achievements, challenges, les-

sons learned, and future steps for the

country, especially in relation to schisto-

somiasis and STHs, are discussed. The

onchocerciasis programme coordinated by

the national programme is not described

in this paper. Details on trachoma that

have not been reported already by others

[11] are presented as part of the ongoing

activities of the NTD programme, as well

as the mapping results for LF.

Assessing the burden of NTDsNTD mapping was an important step

for defining the population at risk of

infections and for rolling out drug delivery

strategies at the national level.

In 2007, 20 primary schools in the

country were randomly selected to assess

Citation: Ndayishimiye O, Ortu G, Soares Magalhaes RJ, Clements A, Willems J, et al. (2014) Control ofNeglected Tropical Diseases in Burundi: Partnerships, Achievements, Challenges, and Lessons Learned afterFour Years of Programme Implementation. PLoS Negl Trop Dis 8(5): e2684. doi:10.1371/journal.pntd.0002684

Editor: William Evan Secor, Centers for Disease Control and Prevention, United States of America

Published May 1, 2014

Copyright: � 2014 Ndayishimiye et al. This is an open-access article distributed under the terms of theCreative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in anymedium, provided the original author and source are credited.

Funding: The NTD programme in Burundi was supported by the investment group Legatum and GenevaGlobal. The funders had no role in study design, data collection and analysis, decision to publish, or preparationof the manuscript.

Competing Interests: The authors have declared that no competing interests exist.

* E-mail: gortu@imperial.ac.uk

PLOS Neglected Tropical Diseases | www.plosntds.org 1 May 2014 | Volume 8 | Issue 5 | e2684

intestinal schistosomiasis and STHs (using

the Kato-Katz method for diagnosis) [12]

and urinary schistosomiasis (using the

urine filtration method). For LF mapping,

because transmission was known to be

minimal at altitudes over 1,300 m [12,13],

the NTD control programme mapped LF

only in communes situated below

1,500 m. A total of 1,754 adults (aged 15

and above with roughly 100 people per

community for a total of 17 communities)

were tested via the LF immunocromato-

graphic test (ICT) card [14].

Results of the parasitological surveys

confirmed that urinary schistosomiasis and

LF were not endemic in the country,

whereas intestinal schistosomiasis was

highly focalized in areas with close proximity

to water bodies. STHs, particularly hook-

worm, were widely distributed, supporting

a mass drug treatment approach targeting

nearly the whole population [15]. The

enclosed supporting information (Tables

S1, S2, S3) outlines the estimated preva-

lence of worm infections in each commune,

with the corresponding risk category as per

WHO guidelines.

For trachoma, a rapid assessment was

performed [16–18], in which an average

of 50 adults and 50 children aged 1–9

years per community were examined

(n = 2,253 children, n = 1,845 adults). This

survey revealed the presence of active

trachoma in the country and was followed

by a more detailed survey in 2009–2010,

in which 20,659 children were examined

in 11 districts [11]. Results from this

survey demonstrated a prevalence of

children with follicular trachoma (TF)

above 10% in three districts (Buhiga,

Muyinga, and Nyabikere). An extension

of this survey to ten other districts also

showed TF above 10% in the Rutana

district (2013 stakeholders meeting held in

Burundi, unreferenced).

All the data obtained were compiled

and integrated with environmental and

country demographic information to cre-

ate maps predicting areas at risk of

infections (Tables S1, S2, S3) [15] that

were then used for defining drug treatment

strategies for each mapped disease. Figure 1

shows the coendemicity by commune for

trachoma and onchocerciasis in 2010 and

Figure 1. Coendemicity map in Burundi communes in 2010 for trachoma, onchocerciasis, schistosomiasis, and soil-transmittedhelminth infections. Coendemicity in Burundi after results obtained from mapping surveys performed in 2007–2010. Colours and patterns indicatepresence of disease in cases of onchocerciasis or prevalence of TF below 10% (low) or above $10% (high) in cases of trachoma. For schistosomiasisand soil-transmitted helminth infections, colours and patterns indicate disease prevalence, which was estimated via predictive risk maps created withthe epidemiological and geospatial integrated approach. In detail, for schistosomiasis, low, moderate, and high indicate prevalence ,10%, between10% and 50%, and above 50%, respectively. For soil-transmitted helminth infections, low, moderate, and high indicate prevalence below 20%,between 20% and 50%, and above 50%, respectively. Abbreviations: Schisto, schistosomiasis; Oncho, onchocerciasis.doi:10.1371/journal.pntd.0002684.g001

PLOS Neglected Tropical Diseases | www.plosntds.org 2 May 2014 | Volume 8 | Issue 5 | e2684

the areas estimated at risk of SCH and

STHs as per mapping results finalized in

2008. Areas with predicted prevalence

estimates $10% for SCH, any STH, or

TF were considered at risk of infection.

Based on these results and on the census

performed in 2008 [19], the estimated

population at risk was calculated based on

an annual growth of 1.03%. Children

between 1 and 4 years of age and 5 and

14 years of age and pregnant women

between the ages of 15 and 49 represented

14.2%, 26.7%, and 5.0% of the total

population, respectively, as estimated by

the census. Two thirds of the pregnant

women were assumed to be in the 2nd or

3rd trimester of pregnancy.

Programme implementationPartnerships and leadership. In

2007 the Burundian MoH developed a

national NTD working plan and

partnered with various organisations for

implementation of a four-year programme

(2007–2011) (Figure 2). Funding was

provided by the investment group

Legatum [20] and was channelled

through Geneva Global [21], the

international philanthropy adviser to the

Global Network for NTD Control

(GNNTDC) [22], which acted as the

primary awarder. The Schistosomiasis

Control Initiative (SCI) [23,24], an NGO

which has been running NTD control

programmes in Africa since 2003, used its

expertise to assist the Burundian MoH

NTD control programme in developing

action plans. CBM [4], an international

NGO that works on preventing and curing

disabilities, was subcontracted to assist the

NTD control programme by providing

management support.

The MoH provided 20 civil servants

who constituted the NTD team and office

space. The newly formed NTD group

represented a focal point within the MoH

for consolidating partnerships with other

international organizations. These new

partnerships facilitated donation of drugs

such as MBZ and/or albendazole (ALB)

from Feed the Children International,

UNICEF, and Food for the Hungry.

PZQ was purchased by SCI on behalf of

Geneva Global, whereas a partnership

with the International Trachoma Initiative

led to the donation of azithromycin (AZT).

Drug distribution strategies:

Options and opportunities. In 2007,

the existing CDTI structure had been

established in six districts endemic for

onchocerciasis, in Rutana, Bururi, and

Makamba provinces. In contrast, the

MCHW strategy was a national

campaign, occurring twice per year, in

which vitamin supplementation,

vaccinations, and MBZ were distributed

to preschool-age children at health posts

available throughout the country.

Based on these health structures and the

information on NTD coendemicity ob-

tained via the mapping surveys, the MoH

piloted an integrated drug distribution in

December 2007 in which people in areas

endemic for onchocerciasis, SCH, and

STHs received IVM, PZQ, and ALB.

However, the integrated delivery strategy

that used Community Directed Distribu-

tors (CDDs) had to be reviewed for the

following reasons: (a) distribution of addi-

tional drugs was perceived as an increased

workload by volunteer CDDs; hence,

there was a risk that extra incentives

would be requested for delivering addi-

tional drugs, and (b) the CDTI network

only covered 6 districts at that time,

whereas some drugs, such as ALB, had

to be distributed at the national level. The

creation of a CDTI network for the entire

country was not an option as it would have

taken many years and the population

needed to be treated as soon as possible.

The alternate system that was used in

the country, the MCHW, performed

better in terms of geographic coverage.

However, both drug delivery systems,

CDTI and MCHW, did not have school-

age children (SAC) as the primary target

population in the treatment campaigns.

Children between 5 and 14 years of age

are the most affected by schistosomiasis

and to some extent by STHs, and

alternative strategies had to be designed

to specifically target this population. One

option was to use primary schools as drug

distribution points to ensure that SAC

were treated in schools during treatment

campaigns. For SAC not enrolled in

primary schools, possible options were to

make PZQ and ALB available in health

posts where treatments for children (aged

under 5) were already distributed and to

create extra distribution points in areas in

which health posts were far from villages.

To accomplish these treatment goals, the

Ministry of Health and the Ministry of

Education agreed to endorse PZQ and

ALB administration to children in primary

schools during the MCHW. During the

national campaigns, mobile clinics (‘‘sites

avances’’) were also set up to specifically

target SAC that were not enrolled in

schools and for villages located more than

5 km from health posts.

In 2008, the combination of MCHW,

primary school treatments, and mobile

clinics was embraced as the best-perform-

ing strategy for delivering nationwide

antihelmintic treatments to children 1- to

4-years-old (MBZ), SAC (PZQ and ALB),

and mothers (ALB). In areas endemic for

trachoma, azithromycin and tetracycline

distribution was done through health

workers in villages and in health centres.

Distribution of these drugs was not

integrated with other NTD treatment

campaigns as areas at risk of trachoma

were limited; based on the available

epidemiological data, it seemed that only

one round of treatment was required.

Figure 3 summarizes the delivery strategies

adopted in Burundi during the NTD

programme and their target populations.

Rolling out MCHW, school-based,

and mobile clinic treatment. Imple-

mentation of this innovative strategy

involved several steps: (a) setting up a

committee to engage partners in financing

several activities related to the national

mass drug administration campaign and to

identify potential financial gaps; (b)

training on drug administration proce-

dures; (c) mobilizing the target population

with national media, local leaders (e.g.,

priests and imams), and village chiefs;

(d) dispatching drugs to the district

and commune levels as well as to

schools; (e) launching the mass drug

administration (MDA); (f) supervising

drug distribution activities to monitor

shortages and verify completion of WHO

template treatment forms [25]; and (f)

evaluating the national programme and

geographical drug coverage as per WHO

guidelines [26].

On a daily basis in each district, the

number of people whose treatment was

observed was entered in a database that

was sent to a provincial office and finally

reported to a central office in charge of

combining data from all 45 districts.

Programme managers set an expectation

of reaching 25% coverage on the first day

of the campaign. If coverage was below

this expectation, village chiefs were

alerted and solicited to sensitize the

population.

Trachoma treatment and PZQ,

IVM, and ALB administration in

adults. Treatment for trachoma was

limited to those cases that were found

positive after the rapid assessment

performed in year 2 and the survey

performed in year 3. In year 4, the

programme distributed azithromycin and

tetracycline (TRT) in three of the four

endemic districts (Buhiga, Muyinga, and

Nyabikere) where TF was found above

10%. A total of 668,719 people were

treated, reaching a programme coverage

of 91% (calculated with the whole

population as target) and a geographic

coverage of 75%. For Rutana district,

PLOS Neglected Tropical Diseases | www.plosntds.org 3 May 2014 | Volume 8 | Issue 5 | e2684

treatment was provided at a later stage

(September 2012).

During the NTD programme, CDDs

continued IVM treatment in areas with

onchocerciasis. In 2007, during the at-

tempt to integrate preventive chemother-

apy treatment for onchocerciasis, schisto-

somiasis, and STHs in adults, a

programme coverage of 52% for PZQ

(calculated with the whole population as

target) and 77% for ALB was achieved in

three of the four provinces (Bururi,

Cibitoke, and Bubanza) where these

diseases were coendemic. The supplemen-

tary data files (Tables S1, S2, S3) provide

details of the communes where adults were

treated with PZQ during this campaign.

Fear of secondary effects in the population

naıve to PZQ was possibly the reason for

the low PZQ coverage in this first

integrated administration attempt and for

abandoning this strategy in 2008 and

2009.

In 2010, the MoH decided to reorga-

nize a combined ALB+PZQ+IVM treat-

ment campaign in the same provinces plus

Rutana. Along with IVM distribution, a

total of 621,268 adults and 3,437,164

people were treated with PZQ and ALB,

respectively, reaching programme cover-

age of 114.6% for PZQ and 96% for ALB.

Although PZQ and ALB coverage im-

proved in 2010, the combined ALB+PZQ+IVM treatment approach was used

only for the adult population in the

following years. Treatment of other groups

continued via the MCHW+schools+mo-

bile clinics strategy that had been adopted

during MDA.

ALB and PZQ preventive chemo-

therapy in mass drug treatments.

PZQ administration targeting children

between 5 and 14 years of age was done

once a year in 29 out of 129 communes,

whereas ALB (or MBZ) was administered

twice per year in the whole country (45

districts), targeting 1- to 14-year-old

children and their mothers (Figure 3).

Following this strategy, Burundi per-

formed one MDA in June 2007 and then

one in June and one in December of every

year until May 2011 (Table 1). A total of

31,664,642 million treatments were deliv-

ered in Burundi during MDAs, of which

2,287,704 were SAC treated for schisto-

somiasis and 28,700,232 pre-SAC and

SAC treated for STHs.

Table 2 shows the ALB programme

coverage at the district level in each target

group. Throughout the whole programme,

reported coverage in children was above

95% in the majority of the districts;

however, some of them reported converg-

es greater than 100%. High coverage

reported by the MoH in some districts

may be related to children receiving drugs

at schools as well as health centres during

the same MDA, resulting in double

recording of treatment. Furthermore, the

wave of unregistered refugees from bor-

dering countries in 2008 increased the

total number of people who were treated,

inflating the coverage as they were unlikely

to have been captured in the estimate of

total number at risk.

A total of 676,706 women in the 2nd

and 3rd trimester of pregnancy were given

ALB during the MCHW, reaching pro-

gramme coverage between 35 and 47%

during the campaigns (Tables 1–2). This

low coverage could possibly be explained by

Figure 2. Partners in the NTD control programme in Burundi.doi:10.1371/journal.pntd.0002684.g002

PLOS Neglected Tropical Diseases | www.plosntds.org 4 May 2014 | Volume 8 | Issue 5 | e2684

the fact that during prenatal consultations,

women diagnosed with worm infections

were often advised not to take any drug

during pregnancy. International guidelines

supporting helminth treatment in pregnant

women [27,28] encouraged the government

to offer ALB as a blanket treatment to

pregnant women for free during the

MCHW campaign. However, during this

campaign, women were free to make the

final decision on either taking ALB free of

charge (irrespective of their infection status)

or paying a fee for treatment at the health

centre if infection was found after the

pregnancy was over. Currently the MoH is

looking into new strategies to sensitise

pregnant women to deworming activities.

For PZQ, treatment was limited to 29

communes at risk of infection (Tables S1, S2,

S3). Because treatment decisions were made

at the commune level and the number of

people treated was reported by district, it was

not possible to calculate the geographic

coverage or the programme coverage for

each commune every year. From personal

communications, reported coverage at the

national level for PZQ administration in

school-age children was 87.9%, 92.2%,

114.6%, and 95.9% in MDAs performed in

the years 2008–2011, respectively; however,

it was not clear whether this coverage was

calculated based on the targeted population

at the commune level or at the district level.

Discussion

Opportunities, challenges, andlessons learned

Despite the sociopolitical instability, Bur-

undi has gradually developed a structured

NTD control programme. Figure 4 sum-

marizes the various stages of this pro-

gramme. From two vertical programmes

focusing on onchocerciasis and malaria,

Burundi recognized the importance of

addressing other NTDs affecting the pop-

ulation and the need to integrate their

prevention and control with health care

initiatives already in place in the country,

an approach that has already been proven

by others to be more sustainable in the long

run [29–31]. The combination of the

national Expanded Programme on Immu-

nization via MCHW with deworming drug

delivery, the intersectoral collaboration

with primary schools, and the use of mobile

clinics during the MCHW appeared to be a

very successful strategy that allowed all

target groups to be reached and confirmed

the feasibility of integrating specific health

Figure 3. Drug distribution strategies adopted during the NTD programme for trachoma, onchocerciasis, schistosomiasis, and soil-transmitted helminth infections. Distribution channels, drugs (or drug packages), target groups, and endemic areas where drugs were deliveredare shown in this figure. Children below 1 year of age and women at the first trimester of pregnancy received vaccinations during the MCHW but nopreventive chemotherapy. Adults above 15 years of age in areas at risk of schistosomiasis but not endemic for onchocerciasis (or trachoma) were nottargeted by any of the health campaigns and did not receive PZQ. * Considering that no test was available to confirm pregnancy in the first 3 months,women often could not receive ALB if uncertain about their status.doi:10.1371/journal.pntd.0002684.g003

PLOS Neglected Tropical Diseases | www.plosntds.org 5 May 2014 | Volume 8 | Issue 5 | e2684

Ta

ble

1.

Ind

ivid

ual

str

eat

ed

for

sch

isto

som

iasi

san

dST

Hs

ine

ach

MD

Ain

Bu

run

di.

Ta

rge

tg

rou

p2

00

72

00

8*

20

09

*2

01

02

01

1T

ota

lin

div

idu

als

tre

ate

din

all

MD

As

Jun

eJu

ne

De

cJu

ne

De

cJu

ne

De

cJu

ne

Sch

isto

som

iasi

s(a

ge

gro

up

5–

14

y/o

)

N/A

58

8,2

14

N/A

54

5,9

50

N/A

55

1,7

57

N/A

60

1,7

83

2,2

87

,70

4

ST

Hs

(ag

eg

rou

p1

–4

y/o

)1

,03

7,1

69

99

2,9

81

1,0

58

,03

43

,45

8,0

18

1,0

58

,03

41

,00

5,9

97

3,8

12

,95

81

,16

9,6

69

28

,70

0,2

32

ST

Hs

(ag

eg

rou

p5

–1

4y

/o2

,07

4,1

83

3,1

83

,87

22

,39

3,4

29

2,3

93

,42

92

,51

7,2

75

2,5

45

,18

4

ST

Hs

(pre

gn

an

tw

om

en

—2

nd

/3rd

trim

est

er)

98

,16

91

07

,94

0—

—1

22

,12

41

04

,46

61

23

,36

21

20

,64

56

76

,70

6

To

tal

nu

mb

er

of

ind

ivid

ua

lstr

ea

ted

ea

chy

ea

r

3,2

09

,52

14

,87

3,0

07

3,5

59

,40

34

,00

3,9

68

3,5

73

,58

74

,17

9,4

95

3,9

36

,32

04

,43

7,2

81

31

,66

4,6

42

*In

De

cem

be

r2

00

8an

dJu

ne

20

09

,w

om

en

did

no

tre

ceiv

etr

eat

me

nt.

do

i:10

.13

71

/jo

urn

al.p

ntd

.00

02

68

4.t

00

1

Ta

ble

2.

Pro

gra

mm

eco

vera

ge

for

STH

s.

20

07

a2

00

8b

20

09

b2

01

0c

20

11

d

Ta

rge

tg

rou

pJu

ne

Jun

eD

ec

Jun

eD

ec

Jun

eD

ec

Jun

e

Me

an

SD

Me

an

SD

Me

an

SD

Me

an

SD

Me

an

SD

Me

an

SD

Me

an

SD

Me

an

SD

ST

Hs

(1–

4y

/o)

95

.11

1.4

92

.83

0.0

98

.52

6.6

10

8.0

35

.79

5.6

25

.88

1.8

14

.61

08

.31

6.8

98

.64

1.8

ST

Hs

(5–

14

y/o

)9

8.9

19

.51

60

.04

4.2

11

8.5

30

.51

15

.12

9.6

10

9.1

12

.91

14

.55

1.1

ST

Hs

(pre

gn

an

tw

om

en

—2

nd

/3rd

trim

est

er)

38

.01

3.3

43

.11

8.8

--

--

47

.52

2.3

36

.11

2.5

42

.99

.24

2.0

18

.1

Me

ans

and

stan

dar

dd

evi

atio

ns

(SD

)ar

eca

lcu

late

db

ase

do

nd

istr

ict

pro

gra

mm

eco

vera

ge

rep

ort

ed

by

the

Min

istr

yo

fH

eal

thfo

re

ach

dis

tric

taf

ter

eve

ryM

DA

.P

rog

ram

me

cove

rag

ew

asca

lcu

late

das

the

pro

po

rtio

no

fa

spe

cifi

cg

rou

po

fin

div

idu

als

that

rece

ive

dA

LBo

ver

the

tota

ln

um

be

ro

fin

div

idu

als

wit

hin

that

gro

up

inth

eim

ple

me

nti

ng

un

itd

ee

me

dat

risk

of

STH

infe

ctio

n[2

6].

InD

ece

mb

er

20

08

and

Jun

e2

00

9,

wo

me

nw

ere

no

ttr

eat

ed

.G

eo

gra

ph

icco

vera

ge

for

ALB

was

alw

ays

10

0%

.aIn

20

07

cove

rag

ew

asre

po

rte

db

yp

rovi

nce

(n=

17

).b

Dis

tric

tsN

yab

ike

re,

Gah

om

bo

,B

usi

ni,

and

Vu

mb

id

idn

ot

exi

stas

adm

inis

trat

ive

un

its

in2

00

8–

20

09

(n=

41

).cN

ew

adm

inis

trat

ive

syst

em

incl

ud

ed

45

dis

tric

ts(n

=4

5).

dW

om

en

’sco

vera

ge

was

no

tre

po

rte

din

Jun

e2

01

1fo

rK

ayan

zad

istr

ict.

do

i:10

.13

71

/jo

urn

al.p

ntd

.00

02

68

4.t

00

2

PLOS Neglected Tropical Diseases | www.plosntds.org 6 May 2014 | Volume 8 | Issue 5 | e2684

programmes (such as the NTD control)

with primary health care initiatives. To the

best of our knowledge, only Rwanda

applied a similar drug administration

strategy in those years, when the MoH, in

collaboration with the Access Project [32]

and supported by the same donors [33],

started its NTD prevention and control

programme.

It should be noted that treatment

frequency and areas targeted by the

distribution did not always follow the

standardized international treatment

guidelines [27] or the treatment strategies

suggested by the mapping results. For

instance, for schistosomiasis treatment,

according to the mapping in 2008, 24 out

of 129 communes had a predicted preva-

lence of schistosomiasis between 10 and

50%. This suggested that in those com-

munes SAC and at-risk adults should have

received PZQ every two years. However,

the government opted for a more intense

approach with a higher frequency of

treatment, which seemed to be more

suitable in the national context considering

that (a) MCHW were performed twice a

year so the addition of one or more drugs

did not substantially increase the cost of this

campaign, and (b) poor access to clean

water and sanitation were and still are

major factors contributing to reinfection.

Another example is treatment of adults

in onchocerciasis and schistosomiasis

coendemic areas. This group was treated

with PZQ in coendemic areas in five

communes where the schistosomiasis esti-

mated predicted mean prevalence was

above 10%, as well as in eight of those

in which this prevalence was below 10%.

PZQ administration in these eight com-

munes was not a decision driven by the

schistosomiasis disease burden but by the

CDTI structure already in place, and it

was based more on logistic considerations

than epidemiological reasoning. The risk

of using logistic considerations for drug

distribution may lead, as in this case, to

missing PZQ administration in areas of

the country where only schistosomiasis is

endemic or delivering unneeded PZQ in

areas at low risk of schistosomiasis because

a distribution structure is already in place.

Considering that the global PZQ produc-

tion is based on the number of people in

high-risk areas in need of treatment, drug

delivery in areas at low risk of infection

might adversely affect PZQ availability for

priority groups and people at high risk of

schistosomiasis.

Finally, areas identified for PZQ distri-

bution were also adjusted according to the

current changing context. Fifteen com-

munes with a predicted mean prevalence

.10% and 14 with prevalence of ,10%

(Tables S1, S2, S3) were also treated with

PZQ, leaving nine communes that were at

medium predicted risk with no treatment

based on the improved urban setting of

these communes (e.g., the relatively good

socioeconomic conditions) and environ-

mental factors such as the absence of the

snail host and stagnant water.

For STHs, the MoH decided again to

administer more ALB than international

guidelines suggest: seven, ten, and 28

districts out of 45 had a predicted

prevalence above 50%, between 20 and

50%, and between 10 and 20%, respec-

tively (Tables S1, S2, S3). Based on these

estimates and on the WHO guidelines

available at that time [27], the country

should have delivered ALB in only seven

districts twice a year and ten districts once

a year, with other districts adopting a case

management approach.

The programme presented challenges

from which we have learned important

lessons. For instance, the programme did

not include a thorough supervision at the

lowest implementation level: NTD educa-

tional material that reached health units

did not reach primary schools; likewise,

laboratory technicians in decentralized

health centers that were trained in the

Kato Katz method for detection of worm

infections never received the material for

running this test. Better supervision at the

lowest level of the programme implemen-

tation would have ensured that the

primary beneficiaries were properly

reached and that the main goals of these

activities (education on NTDs for SAC in

the first case and detection and treatment

of infected people in the second example)

would have been achieved.

Finally, Burundi identified the com-

mune, and not the district, as the imple-

menting unit for distributing PZQ. Al-

though this strategy appeared to be very

appropriate, reported coverage calculated

at the district level did not allow a rapid

detection of communes that could have

been missed by the MDA or perhaps did

not require MDA. It would have been

more effective to report programme cov-

erage according to the actual implement-

ing unit (commune) to verify in real time

possible mistakes in drug distribution and

implement the correct measures.

Figure 4. Evolution of the NTD prevention and control programme in Burundi.doi:10.1371/journal.pntd.0002684.g004

PLOS Neglected Tropical Diseases | www.plosntds.org 7 May 2014 | Volume 8 | Issue 5 | e2684

Looking aheadBurundi benefitted greatly from this

NTD programme, which has resulted in

the mapping of the four major NTDs

(schistosomiasis, STHs, LF, and partially

trachoma), delivery of treatments reaching

the population at risk of infections with

good coverage, and a five-year national

plan for NTD prevention and control

endorsed by WHO and validated by the

MoH. The programme has also demon-

strated that the integration of NTD

treatments with other health initiatives,

such as the Expanded Programme of

Immunization and the use of mobile

clinics plus primary schools as additional

distribution sites, is a powerful strategy for

reaching the whole population at risk.

With the current drug donation programs

[34,35] and the continuous financial and

technical support from international

NGOs, especially from UNICEF for the

MCHW campaigns, Burundi is in a

position to complete the mapping of

trachoma in the remaining 24 districts

and to continue MDAs for the major

NTDs affecting the population.

For the sustainability of this NTD

programme and with the final aim of

eliminating these diseases, the initiation of

a multidisciplinary approach involving

other ministries, such as those responsible

for water and sanitation, is paramount. All

these intersectoral initiatives could be

effectively coordinated by the government

to ensure their long-term sustainability.

Coordination and effective management

of intersectoral initiatives could be

achieved by establishing technical groups

within the MoH that would be responsible

for linking specific NTD control needs

with other initiatives relevant to this

programme (e.g., water and sanitation

initiatives). Delegation of these groups for

all technical aspects of the NTD manage-

ment and divulgation of intersectoral work

plans and technical material would be the

next steps to achieve an effective coordi-

nation of multidisciplinary activities.

These steps will imply a shift towards

complete ownership of the NTD pro-

gramme by the MoH.

Meanwhile, effort should be put into the

development of an efficient decentralized

routine surveillance system within the

primary health care structure for effective

case detection, management, and follow-

up care. This step should be taken in

Burundi when disease prevalence has been

reduced and case management replaces

mass drug administration. After initial

technical and financial support from

international stakeholders for building

NTD surveillance capacity, financial com-

mitment of the MoH to maintain this

structure in decentralized areas will ulti-

mately be expected for sustainability of the

NTD prevention, control, and surveillance

program in Burundi.

Supporting Information

Table S1 Burundi administrativeunits in year 2012. Administrative units

in Burundi updated to year 2012 (prov-

inces, districts, and communes).

(XLSX)

Table S2 Estimated risk prevalenceof schistosomiasis in each communeand treatment by commune in 2007–2011. Epidemiological data obtained in

2007 during the mapping were compiled

and integrated with environmental and

country demographic information to obtain

estimated mean prevalence of schistosomi-

asis in each commune. Based on WHO

guidelines (2006), communes were classified

as high, moderate, and low risk of infection.

These data were then used to create maps

predicting areas at risk of schistosomiasis

(see Figure 1). The table also provides

information about praziquantel treatment

and the population targeted in each com-

mune.

(XLSX)

Table S3 Estimated risk preva-lence of soil-transmitted helminthinfections in each district. Epidemio-

logical data obtained in 2007 during the

mapping were compiled and integrated

with environmental and country demo-

graphic information to obtain estimated

mean prevalence of soil-transmitted hel-

minth infections in each district. Based on

WHO guidelines (2006), districts were

classified as high, moderate, and low risk

of infection. These data were then used to

create maps predicting areas at risk of soil-

transmitted helminth infections (see

Figure 1). The table also provides infor-

mation about suggested drug strategies for

each district based on predicted risk and

on WHO guidelines available in 2006.

(XLSX)

Acknowledgments

The authors would like to thank the team of the

Burundian NTD Control programme and its

partners: SCI, CBM, the GNNTDC, and

Geneva Global. Finally, our gratitude is extend-

ed to the health staff, teachers, and communities

without whom this programme could not be

successful.

References

1. World Health Organization (2012) Accelerating

work to overcome the global impact of neglected

tropical diseases—A roadmap for implementa-

tion. Available: http://whqlibdoc.who.int/hq/

2012/WHO_HTM_NTD_2012.1_eng.pdf. Ac-

cessed 1 April 2014.

2. Hotez PJ, Molyneux DH, Fenwick A, Ottesen E,

Ehrlich Sachs S, et al. (2006) Incorporating a

Rapid-Impact Package for Neglected Tropical

Diseases with Programs for HIV/AIDS, Tuber-

culosis, and Malaria. PLoS Med 3: e102.

3. World Health Organization (2014) African Pro-

gramme for Onchocerciasis Control (APOC).

Available: http://www.who.int/blindness/

partnerships/APOC/en/. Accessed 1 April 2014.

4. CBM (2014) CBM – ex Christoffel-Blindmission.

Available: http://www.cbmuk.org.uk/. Accessed

1 April 2014.

5. World Health Organization (2012) African Pro-

gramme for Onchocerciasis Control (APOC)—

Community-directed treatment with ivermectin

(CDTI). Available: http://www.who.int/apoc/

cdti/en/. Accessed 1 April 2014.

6. Engels D, Ndoricimpa J, Gryseels B (1993)

Schistosomiasis mansoni in Burundi: progress in

its control since 1985. Bull World Health Organ

71: 207–214.

7. Engels D, Ndoricimpa J, Nahimana S, Gryseels B

(1994) Control of Schistosoma mansoni and

intestinal helminths: 8-year follow-up of an urban

school programme in Bujumbura, Burundi. Acta

Tropica 58: 127–140.

8. World Health Organzation (2005) Communicable

disease toolkit - Burundi Profile. WHO/CDS/

NTD/DCE/2005.2a. Available: http://www.

who.int/diseasecontrol_emergencies/toolkits/

Burundi_profile_ok.pdf. Accessed 1 April 2014.

9. Dismas B (2005) Rapport d’enquete sur la

situation epidemiologique de la bilharziose dans

les ecoles primaires des provinces endemiques du

Burundi. Burundi: Programme de Lutte contre

les Maladies Transmissibles et Carentielles, Min-

istere de la Sante Publique; Burundi. 44 p.

10. Price EW (1976) Endemic elephantiasis of the

lower legs in Rwanda and Burundi. Trop Geogr

Med 28: 283–290.

11. Ndayishimiye O, Willems J, Manirakiza E, Smith

JL, Gashikanyi R, et al. (2011) Population-based

survey of active trachoma in 11 districts of

Burundi. Ophthalmic Epidemiol 18: 146–149.

12. Gyapong JO, Remme JH (2001) The use of grid

sampling methodology for rapid assessment of the

distribution of bancroftian filariasis. Trans R Soc

Trop Med Hyg 95: 681–686.

13. Gyapong M, Gyapong JO, Owusu-Banahene G

(2001) Community-directed treatment: the way

forward to eliminating lymphatic filariasis as a

public-health problem in Ghana. Ann Trop Med

Parasitol 95: 77–86.

14. World Health Organization (2005) Monitoring

and epidemiological assessment of the pro-

gramme to eliminate lymphatic fi lariasis at

implementation unit level. WHO/CDS/CPE/

CEE/2005.50. Available: http://whqlibdoc.who.

int/hq/2005/who_cds_cpe_cee_2005.50.pdf.

Accessed 1 April 2014.

15. Clements AC, Deville MA, Ndayishimiye O,

Brooker S, Fenwick A (2010) Spatial co-distribu-

tion of neglected tropical diseases in the east

African great lakes region: revisiting the justifica-

tion for integrated control. Trop Med Int Health

15: 198–207.

16. World Health Organization (2001) Guidelines for

rapid assessment for bliding trachoma. WHO/

PBD/GET/00.8. Available: http://www.who.

PLOS Neglected Tropical Diseases | www.plosntds.org 8 May 2014 | Volume 8 | Issue 5 | e2684

int/blindness/TRA-ENGLISH.pdf. Accessed 1

April 2014.17. Thylefors B, Dawson CR, Jones BR, West SK,

Taylor HR (1987) A simple system for the

assessment of trachoma and its complications.Bull World Health Organ 65: 477–483.

18. Jeremiah Ngondi MR, Fiona Matthews, CarolBrayne, Paul Emerson (2009) Trachoma survey

methods: a literature review. Bull World Health

Organ 87: 143–151.19. Ministry of Health, Burundi, Burundi (2010)

Troisieme recensement general de la populationet de l’habitat de 2008. Available: http://www.

paris21.org/sites/default/files/BURUNDI-population-2008.PDF. Accessed 1 April 2014.

20. Legatum Foundation (2014) Legatum Founda-

tion. Available: http://www.legatum.com/foundation. Accessed 1 April 2014.

21. Geneva Global (2012) GENEVA GLOBAL—Performance philanthropy. Available: http://

www.genevaglobal.com/. Accessed 1 April 2014.

22. Sabin Vaccine Institute (2014) GNNTD GlobalNetwork of Neglected Tropical Diseases. Avail-

able: http://www.globalnetwork.org/. Accessed 1April 2014.

23. Fenwick A, Webster JP, Bosque-Oliva E, Blair L,Fleming FM, et al. (2009) The Schistosomiasis

Control Initiative (SCI): rationale, development

and implementation from 2002–2008. Parasitol-

ogy 136: 1719–1730.24. Garba A, Toure S, Dembele R, Boisier P, Tohon

Z, et al. (2009) Present and future schistosomiasis

control activities with support from the Schisto-somiasis Control Initiative in West Africa. Para-

sitology 136: 1731–1737.25. World Health Organization (2011) Helminth

control in school-age children. A guide for

programme managers of control programmes.A v a i l a b l e : h t t p : // w h q l i bd o c . w h o . i n t /

publications/2011/9789241548267_eng.pdf. Ac-cessed 1 April 2014.

26. World Health Organization (2010) Monitoringdrug coverage for preventive chemotherapy.

A v a i l a b l e : h t tp :/ / w h q l ib do c . w h o . i n t /

publications/2010/9789241599993_eng.pdf. Ac-cessed 1 April 2014.

27. World Health Organization (2006) Preventivechemotherapy in human helminthiasis. Available:

http://whqlibdoc.who.int/publications/2006/

9241547103_eng.pdf. Accessed 1 April 2014.28. World Health Organization (1994) Informal

consultation on hookworm infection and anaemiain girls and women. WHO/CTD/SIP/96.1.

Available: http://whqlibdoc.who.int/hq/1996/WHO_CTD_SIP_96.1.pdf. Accessed 1 April

2014.

29. Utzinger J, Raso G, Brooker S, De Savigny D,

Tanner M, et al. (2009) Schistosomiasis andneglected tropical diseases: towards integrated

and sustainable control and a word of caution.

Parasitology 136: 1859–1874.30. Zhang Y, MacArthur C, Mubila L, Baker S

(2010) Control of neglected tropical diseasesneeds a long-term commitment. BMC Med 8: 67.

31. Dembele M, Bamani S, Dembele R, Traore MO,

Goita S, et al. (2012) Implementing preventivechemotherapy through an integrated National

Neglected Tropical Disease Control Program inMali. PLoS Negl Trop Dis 6: e1574.

32. ACCESS (2009) Project Access—Annual Report.Available: http://tap.pmhclients.com/images/

u p l o a d s / p r e s s / A N N U A L _ R E P O R T _

19FEB10b.pdf. Accessed 1 April 2014.33. Ruxin J, Negin J (2012) Removing the neglect from

neglected tropical diseases: the Rwandan experi-ence 2008–2010. Glob Public Health 7: 812–822.

34. Gustavsen KM, Bradley MH, Wright AL (2009)

GlaxoSmithKline and Merck: private-sector col-laboration for the elimination of lymphatic filari-

asis. Ann Trop Med Parasitol 103 Suppl 1: S11–5.35. MerckGroup (2012) Global Health Progress.

Available: http://www.globalhealthprogress.org/programs/merck-praziquante l -donat ion-

program. Accessed 1 April 2014.

PLOS Neglected Tropical Diseases | www.plosntds.org 9 May 2014 | Volume 8 | Issue 5 | e2684